Designing choice experiments to Test for Anchoring and Framing Effects

Choice experiments (CE) are increasingly used as a stated preference technique to value changes in non-market goods. Respondents to a CE survey are asked to make repeated choices between alternatives. Each alternative is described by a number of attributes – the attributes levels vary across alternatives and choice sets. A monetary attribute is typically included so that marginal values for changes in the non-market attributes presented can be estimated. The monetary attribute has central importance. However, there has been limited research on the impacts on respondents’ choices of changing the (range in) levels of the monetary attribute presented in CE surveys. This is known as the ‘anchoring’ effect. The ‘framing’ of non-market attributes may also affect value estimates. Attribute framing refers to the context in which the attributes are presented to respondents in a CE survey. The challenge for CE practitioners is to identify how particular attribute frames may influence respondents’ choices. This research report provides a review of anchoring and framing effects in CEs. A CE questionnaire is described to incorporate tests for anchoring and framing effects. Ten hypotheses are developed about the impacts of various attribute ‘anchors’ and ‘frames’ on respondents’ choices and subsequent values estimated.